Package: qrnn 2.1.1

Alex J. Cannon

qrnn: Quantile Regression Neural Network

Fit quantile regression neural network models with optional left censoring, partial monotonicity constraints, generalized additive model constraints, and the ability to fit multiple non-crossing quantile functions following Cannon (2011) <doi:10.1016/j.cageo.2010.07.005> and Cannon (2018) <doi:10.1007/s00477-018-1573-6>.

Authors:Alex J. Cannon [aut, cre]

qrnn_2.1.1.tar.gz
qrnn_2.1.1.tar.gz(r-4.5-noble)qrnn_2.1.1.tar.gz(r-4.4-noble)
qrnn_2.1.1.tgz(r-4.4-emscripten)qrnn_2.1.1.tgz(r-4.3-emscripten)
qrnn.pdf |qrnn.html✨
qrnn/json (API)

# Install 'qrnn' in R:
install.packages('qrnn', repos = c('https://cran.r-universe.dev', 'https://cloud.r-project.org'))

Peer review:

Datasets:
  • YVRprecip - Daily precipitation data at Vancouver Int'l Airport

This package does not link to any Github/Gitlab/R-forge repository. No issue tracker or development information is available.

3.95 score 8 stars 1 packages 74 scripts 666 downloads 43 exports 0 dependencies

Last updated 10 months agofrom:61ef772fff. Checks:OK: 2. Indexed: yes.

TargetResultDate
Doc / VignettesOKDec 01 2024
R-4.5-linuxOKDec 01 2024

Exports:adamcensored.meancomposite.stackdquantiledummy.codeeluelu.primegam.stylehramphramp.primehuberhuber.primelinearlinear.primelogisticlogistic.primelrelulrelu.primemcqrnn.fitmcqrnn.predictpquantilepquantile.nwqquantileqquantile.nwqrnn.costqrnn.fitqrnn.initializeqrnn.predictqrnn.rbfqrnn2.fitqrnn2.predictrelurelu.primerquantilerquantile.nwsigmoidsigmoid.primesoftmaxsoftplussoftplus.primetilted.abstilted.approxtilted.approx.prime

Dependencies:

Readme and manuals

Help Manual

Help pageTopics
Quantile Regression Neural Networkqrnn-package qrnn
Adaptive stochastic gradient descent optimization algorithm (Adam)adam
A hybrid mean/median function for left censored variablescensored.mean
Reformat data matrices for composite quantile regressioncomposite.stack
Convert a factor to a matrix of dummy codesdummy.code
Modified generalized additive model plots for interpreting QRNN modelsgam.style
Huber norm and Huber approximations to the ramp and tilted absolute value functionshramp hramp.prime huber huber.prime tilted.approx tilted.approx.prime
Monotone composite quantile regression neural network (MCQRNN) for simultaneous estimation of multiple non-crossing quantilesmcqrnn mcqrnn.fit mcqrnn.predict
Smooth approximation to the tilted absolute value cost functionqrnn.cost
Main function used to fit a QRNN model or ensemble of QRNN modelsqrnn.fit
Initialize a QRNN weight vectorqrnn.initialize
Evaluate quantiles from trained QRNN modelqrnn.predict
Radial basis function kernelqrnn.rbf
Fit and make predictions from QRNN models with two hidden layersqrnn2.fit qrnn2.predict
Interpolated quantile distribution with exponential tails and Nadaraya-Watson quantile distributiondquantile pquantile pquantile.nw qquantile qquantile.nw rquantile rquantile.nw
Tilted absolute value functiontilted.abs
Transfer functions and their derivativeselu elu.prime linear linear.prime logistic logistic.prime lrelu lrelu.prime relu relu.prime sigmoid sigmoid.prime softmax softplus softplus.prime
Daily precipitation data at Vancouver Int'l Airport (YVR)YVRprecip